In pilot studies we have demonstrated DNA content abnormalities in both uveal melanomas (p less than 0.0003) and retinoblastomas (p less than 0.05), that correlate with prognosis. Using in vivo BrdUrd assays, we have observed that uveal melanoma cell cycling is decreased after either 20 Gy of pre-enucleation photon irradiation or greater than 60 GyE of helium ion therapy (p less than 0.0001). We wish to perform a number of archival and prospective studies to determine the accuracy and sensitivity of these techniques to determine both prognosis and radiation efficacy in uveal melanoma patients. Fine needle aspiration biopsy from non-enucleated and enucleated eyes will be compared with whole tumor tissue for both BrdUrd cell cycling studies and DNA content analysis. We will prospectively determine whether these assays will improve our abilities to predict patient survival and radiation response, and measure the relative specificities and sensitivities of two color image analysis, flow cytometry, and standard immunofluorescent microscopy measurement of cell cycle and DNA content abnormalities. We will also compare the prognostic accuracy of standard deviation of nucleolar area (SDNA) with DNA content analysis in a masked retrospective uveal melanoma study. These findings have a number of important ramifications, and require in depth analysis. Many uveal melanomas are now managed without enucleation, and classic histopathologic prognostic factors are not available. We will determine if DNA content abnormalities, especially on fine needle biopsy, can be used to increase prognostic accuracy. It is often difficult to determine whether radiation has been effective in treatment of a uveal melanoma. It is likely that assessment of cell cycle status both before, and in some cases after radiation may improve our ability to determine whether a tumor's reproductive integrity has been destroyed, especially in cases where there has been questionable growth after either 125I radioactive plaque or helium ion therapy. These studies will also serve as an excellent model to determine the relative advantages and disadvantages of flow cytometry and two-color image analysis for cell cycle and DNA content investigations.